Store-Independent Orai Channels Regulated by STIM

Excerpt

The identification of Orai and STIM proteins has opened up new avenues of research in the field of receptor-regulated calcium signaling. The ligation of phospholipase C (PLC)-coupled receptors can activate either the common store-operated calcium entry (SOCE) pathway or store-independent calcium entry (SICE) pathway. The representative conductance of the SOCE pathway is the calcium release-activated calcium (CRAC) channel encoded by Orai (CRACM) proteins. The SICE pathway biophysical manifestation is currents activated by arachidonic acid (AA) or the AA metabolite leukotriene C₄ (LTC₄) and termed arachidonate-regulated or LTC₄-regulated calcium (ARC/LRC) current encoded by channels composed of both Orai1 and Orai3 proteins.

About three decades ago, Putney first proposed the capacitative Ca²⁺ entry model (subsequently known as SOCE) [1]. Orai1 protein, the pore forming subunit of the CRAC channel, was discovered almost simultaneously by three groups in 2006 [2–4]. The Orai family of channels contains three different proteins (Orai1, 2, and 3) encoded by independent genes [5]. A large number of agonists can act on G protein-coupled receptors (GPCRs) to activate PLC. PLC hydrolyzes phosphatidylinositol-4,5-bisphopshate (PIP₂) into diacylglycerol (DAG) and inositol-1,4,5-trisphosphate (IP₃) [6]. The latter binds to IP₃ receptors (IP₃R) on the membrane of the endoplasmic reticulum (ER), resulting in Ca²⁺ store emptying. The action of ER Ca²⁺ store emptying causes stromal interaction molecule 1 (STIM1), a calcium sensor to lose Ca²⁺ from its N-terminal low affinity EF hand located in the lumen of the ER [7,8]. This causes STIM1 to aggregate and to move to highly specialized areas where the ER comes close to the plasma membrane to physically trap and interact with Orai1 channels and activate Ca²⁺ entry [8]. STIM1 has one homologue, STIM2, which mediates Orai1 channel activation under resting conditions in the absence of agonist stimulation [9]. In most cells studied so far, SOCE is mediated by STIM1 and Orai1 proteins [10]. However, we reported Orai3-mediated SOCE in a subset of estrogen receptor positive breast cancer cells [11–13].

The ARC channel and its role in calcium signaling have been first reported and intensely studied by Shuttleworth and colleagues [14–17]. This group identified and characterized a conductance in HEK293 cells activated by relatively low exogenous concentrations of arachidonic acid or by low concentrations of a muscarinic agonist [18]. Polyunsaturated fatty acids were described as poor activators of these channels, and mono unsaturated or saturated fatty acids were ineffective [19]. Unlike a number of channels of the Transient Receptor Potential Canonical (TRPC3/6/7) family [20,21], ARC channels are not activated by high concentrations of DAG (100 μM) [19]. Mignen and colleagues showed that despite many similarities with CRAC currents present in the same cell type studied, ARC channels possess distinct pharmacological characteristics and biophysical properties [18]. For instance, unlike CRAC channels, ARC channels do not show the typical fast Ca²⁺-dependent inactivation (CDI), are not inhibited by a reduction in extracellular pH from 7.2 to 6.7, and are insensitive to 2-aminoethoxydiphenyl borate (2-APB) [18,22]. As is the case with CRAC channels [34,35], the absence of divalent cations in the extracellular recording medium induces the permeability of ARC channels to monovalent cations, such as Na⁺ [23]. However, this monovalent macroscopic current has different characteristics from those observed for CRAC channels especially from the perspective of their depotentiation and permeability. By blocking monovalent currents by increasing extracellular calcium concentrations as a relative measure of selectivity of calcium channels, Mignen and colleagues proposed that ARC channels have high Ca²⁺ selectivity and are 50 times more Ca²⁺-selective than CRAC channels [18,22]. These authors argued that ARC channels are the predominant calcium channels activated when cells are stimulated with low concentrations of agonists that induce repetitive calcium oscillations [24]. Using an M3 muscarinic receptor-expressing HEK293 cells and murine parotid and pancreatic acinar cells, they reported the activation of ARC channels mediating intracellular calcium oscillations by low concentrations (0.2–1 μM) of carbachol [25]. In the same cells they described the activation of the AA-producing enzyme, phospholipase A2 type IV, upon stimulation with low concentrations of carbachol.

Earlier work by the Shuttleworth group suggested that the pharmacological inhibition of PLA2 with isotetrandrine blocks the activation of ARC channels, while the pharmacological inhibition of the lipoxygenase and cyclooxygenase pathways had no effect on ARC activation [14,26], indicating that AA is produced by receptor-mediated activation of PLA2 and that AA processing into downstream metabolites is not required for ARC channel activation. After identification of STIM and Orai proteins, Shuttleworth and colleagues showed that both Orai1 and Orai3 are required for ARC channel activation [27], in addition to the minor pool of STIM1 located in the plasma membrane [28]. More recent work from our laboratory identified a SICE channel in primary aortic vascular smooth muscle cells (VSMC). We found that this conductance is activated by AA, but AA metabolism into LTC₄ by the enzymatic activity of LTC₄ synthase (LTC₄S) provided a more robust activation of these channels; LTC₄ acts intracellularly when applied through the patch pipette but not extracellularly when added to the bath solution. We named this channel LTC₄-regulated calcium (LRC) channel [23,29–31]. Collectively, our data in VSMC showed that receptor activation causes production of AA through sequential activation of PLC and DAG lipase and that AA metabolism by 5-lipooxygenase and LTC₄S into LTC₄ is required for LRC channel activation [29,31]. A molecular knockdown on LTC₄ synthase (LTC₄S) abrogated receptor-mediated LRC channel activation (using the PAR1 agonist thrombin), while direct application of LTC₄ through the patch pipette robustly activated LRC currents. The biophysical properties of LRC channels were identical to those of ARC channels, prompting us to undertake a side by side comparison in VSMC and HEK293 cells to determine whether these two conductances are mediated by the same or by different cellular pools of STIM and Orai proteins [23]. Briefly, using protein knockdown, pharmacological inhibitors, and a nonmetabolizable form of AA, we found that regardless of the cell type considered (HEK293 cells or VSMC), ARC and LRC currents are the manifestation of the same channel that can be activated by AA but is more robustly activated by LTC₄ [23]. We also found that in both cell types, ARC/LRC currents depended on Orai1, Orai3, and STIM1 [23,29], but unlike findings from the Shuttleworth group, we were able to rescue ARC/LRC activity in HEK293 cells and VSMC with expressed STIM1 constructs that do not traffic to the plasma membrane when using Fura-2 calcium imaging and perforated patch recording in intact cells but not in whole-cell recordings. These results suggest a facilitatory role for PM-STIM1 in ARC/LRC channel activation [23]. Orai1 exists in two variants generated through alternative translation-initiation of the Orai1 mRNA: a longer Orai1α form contains an additional N-terminal (NT) 63 amino acids upstream of the conserved start site of a shorter Orai1β [32]. A study from our group showed that while Orai1α and Orai1β are interchangeable for forming CRAC channels, only Orai1α can support ARC/LRC channels by forming a unique heteromeric channel with Orai3. Studies by the Shuttleworth group were performed before Orai1α variant was discovered; it is therefore unclear which Orai1 subtype was used [33]. We also showed that a specific interaction of STIM1 second C-terminal (CT) coiled-coil (CC2) with Orai3 CT region is required for LRC channel activation by LTC₄ [31].

In summary, the SICE pathway appears to be mediated by one channel entity. In succeeding text, we will refer to this channel as either ARC or LRC, depending on whether we are referring to experiments that used either AA or LTC₄ to activate this conductance. ARC/LRC channels are encoded by Orai1 and Orai3 and regulated by STIM1. There are two major points of contention between our findings and those of the Shuttleworth group: (1) the requirement for AA metabolic conversion into LTC₄ and (2) the cellular pool of STIM1 required for ARC/LRC activation, that is, ER-resident versus PM-resident STIM1.